Speakers
Description
With the increasing use of cloud services in critical industrial and public systems, reliability assessment has become an important research and engineering challenge. Traditional approaches such as probabilistic models and Fault Tree Analysis (FTA) provide a solid analytical basis, but they are often limited in representing the dynamics and interdependencies of modern cloud infrastructures. This paper presents a conceptual hybrid framework for reliability assessment in cloud environments that combines probabilistic-statistical analysis, intelligent predictive methods, and scenario-based reasoning. To illustrate the applicability of the proposed framework, an analytical case study based on Fault Tree Analysis is developed for a cloud service failure scenario. The study estimates the probability of a top-level service outage and applies importance measures to identify the most critical contributing factors. The obtained results show how classical reliability analysis can be structured to support future integration with adaptive and data-driven methods. The proposed framework should be viewed as a foundation for subsequent empirical validation with real cloud monitoring data and comparative predictive experiments.